Public Graduation Presentations Friday, November 24
Posted: 17 November 2017 04:01 PM
Total Posts:  185
Joined  2013-12-02

You’re all invited to the Public Graduation Presentations of Max Roele and Sam van Tienhoven on Friday, November 24, 2017.
Location: Academiegebouw room 01, Rapenburg 73, 2311, Leiden.

15.00 - 15.35 Max Roele
15.45 - 16.20 Sam van Tienhoven

15.00 - 15.35 Max Roele
Title:Tweet with a Smile and a Unicorn: emoji usage in the Netherlands and England
Abstract: With the immense growth of social media, emoji have also spread all over the world. Currently, there are more than a thousand of these colourful pictographs showing faces, creatures and objects. However, not all of these emoji see regular use, while others are very popular. This exploratory study attempts to answer the following question: how and why are emoji used on Twitter in the Netherlands and England? Through a quantitative analysis of two million tweets, current emoji usage in the two countries is described. A following qualitative study consisting of sixteen semi-structured interviews provides insight into why individuals use emoji and how they select emoji. Three important themes that influence both the number of emoji used and the selection of emoji were found: the individual’s purpose on Twitter, the perceived functionality of emoji, and the individual’s selection criteria that emoji had to fulfill. Overall, emoji play an important role in online communication and their functions and usage are more complex than their light-hearted appearance suggests.
Thesis advisors: Max van Duijn and Janelle Ward (Erasmus University Rotterdam)

15.45 - 16.20 Sam van Tienhoven
Title:Melodic Swarms: generation of melody in harmonic context using swarm algorithms
Abstract: Artificial intelligence techniques have been applied to music composition in various ways, among which are swarm algorithms. The self organizing property of swarm algorithms is similar to certain musical behavior. Research on applying swarm algorithms to composition has led to systems being able to create complex soundscapes and musical textures. However, the composition of melodies using swarm algorithms has not been properly studied. Using an experimental approach, we created multiple swarm algorithm systems that generate melodic phrases based on a given chord progression. These approaches are then assessed on basis of their individual merits and in comparison to each other. Concretely, we try to distill the specific traits and benefits of using swarm algorithms for the generation of melodies. We then discuss further research and possible applications of melodic swarms.
Thesis advisors: Edwin van der Heide , Daniel Bisig and Philippe Kocher